[C2-devel] question about the clustering algorithm

[Pranav <un...@gm...>]

I'm running the DCS demo app and have tried out the various algorithm
implementations provided. Now, the input to the clustering is a series of
documents of the type:

 String id           // Document #1
 String url
 String title
 String snippet
 String id           // Document #1
 String url
 String title
 String snippet

Do the clustering algorithms give equal emphasis on the 'title' as well as
the 'snippet'..If so, is there a way to force greater emphasis on the title
? The results I'm seeing are getting a bit skewed and would correct a whole
lot if more focus could be placed on the 'title'...

thanks,
-pranav.